Gas-Particle and Granular Flow

Systems: Coupled Numerical Methods

and Applications 1st Edition Nan Gui
Visit to download the full and correct content document:
https://textbookfull.com/product/gas-particle-and-granular-flow-systems-coupled-num
erical-methods-and-applications-1st-edition-nan-gui/
More products digital (pdf, epub, mobi) instant
download maybe you interests ...

Traffic and Granular Flow 15 1st Edition Victor L.

Knoop

https://textbookfull.com/product/traffic-and-granular-
flow-15-1st-edition-victor-l-knoop/

Gas Turbines Internal Flow Systems Modeling Bijay

Sultanian

https://textbookfull.com/product/gas-turbines-internal-flow-
systems-modeling-bijay-sultanian/

Environmental Flow Assessment: Methods and Applications

1st Edition John G. Williams

https://textbookfull.com/product/environmental-flow-assessment-
methods-and-applications-1st-edition-john-g-williams/

Flow Assurance Solids in Oil and Gas Production 1st

Edition Jon Steinar Gudmundsson

https://textbookfull.com/product/flow-assurance-solids-in-oil-
and-gas-production-1st-edition-jon-steinar-gudmundsson/
Multiphysics Modeling: Numerical Methods and
Engineering Applications: Tsinghua University Press
Computational Mechanics Series 1st Edition Cen

https://textbookfull.com/product/multiphysics-modeling-numerical-
methods-and-engineering-applications-tsinghua-university-press-
computational-mechanics-series-1st-edition-cen/

Intelligent numerical methods Applications to

Fractional Calculus George A. Anastassiou

https://textbookfull.com/product/intelligent-numerical-methods-
applications-to-fractional-calculus-george-a-anastassiou/

Single Particle Nanocatalysis Fundamentals and

Applications Chen

https://textbookfull.com/product/single-particle-nanocatalysis-
fundamentals-and-applications-chen/

Functional Numerical Methods: Applications to Abstract

Fractional Calculus 1st Edition George A. Anastassiou

https://textbookfull.com/product/functional-numerical-methods-
applications-to-abstract-fractional-calculus-1st-edition-george-
a-anastassiou/

Light Scattering by Ice Crystals Fundamentals and

Applications 1st Edition Kuo-Nan Liou

https://textbookfull.com/product/light-scattering-by-ice-
crystals-fundamentals-and-applications-1st-edition-kuo-nan-liou/
 Gas-Particle and
Granular Flow Systems
Coupled Numerical Methods
and Applications
 Gas-Particle and
Granular Flow Systems
Coupled Numerical Methods
and Applications
Nan Gui
Institute of Nuclear and New Energy Technology
Collaborative Innovation Center of Advanced Nuclear Energy Technology
Key Laboratory of Advanced Reactor Engineering and Safety
Ministry of Education, Tsinghua University
Beijing, China

Shengyao Jiang
Institute of Nuclear and New Energy Technology
Collaborative Innovation Center of Advanced Nuclear Energy Technology
Key Laboratory of Advanced Reactor Engineering and Safety
Ministry of Education, Tsinghua University
Beijing, China

Jiyuan Tu
Institute of Nuclear and New Energy Technology
Collaborative Innovation Center of Advanced Nuclear Energy Technology
Key Laboratory of Advanced Reactor Engineering and Safety
Ministry of Education, Tsinghua University
Beijing, China
School of Engineering, RMIT University
Melbourne, VIC, Australia

Xingtuan Yang
Institute of Nuclear and New Energy Technology
Collaborative Innovation Center of Advanced Nuclear Energy Technology
Key Laboratory of Advanced Reactor Engineering and Safety
Ministry of Education, Tsinghua University
Beijing, China
Elsevier
Radarweg 29, PO Box 211, 1000 AE Amsterdam, Netherlands
The Boulevard, Langford Lane, Kidlington, Oxford OX5 1GB, United Kingdom
50 Hampshire Street, 5th Floor, Cambridge, MA 02139, United States
Copyright © 2020 Elsevier Inc. All rights reserved.
No part of this publication may be reproduced or transmitted in any form or by any means, electronic
or mechanical, including photocopying, recording, or any information storage and retrieval system,
without permission in writing from the publisher. Details on how to seek permission, further
information about the Publisher’s permissions policies and our arrangements with organizations such
as the Copyright Clearance Center and the Copyright Licensing Agency, can be found at our website:
www.elsevier.com/permissions.
This book and the individual contributions contained in it are protected under copyright by the
Publisher (other than as may be noted herein).
Notices
Knowledge and best practice in this field are constantly changing. As new research and experience
broaden our understanding, changes in research methods, professional practices, or medical treatment
may become necessary.
Practitioners and researchers must always rely on their own experience and knowledge in evaluating
and using any information, methods, compounds, or experiments described herein. In using such
information or methods they should be mindful of their own safety and the safety of others, including
parties for whom they have a professional responsibility.
To the fullest extent of the law, neither the Publisher nor the authors, contributors, or editors, assume
any liability for any injury and/or damage to persons or property as a matter of products liability,
negligence or otherwise, or from any use or operation of any methods, products, instructions, or ideas
contained in the material herein.

Library of Congress Cataloging-in-Publication Data

A catalog record for this book is available from the Library of Congress

British Library Cataloguing-in-Publication Data

A catalogue record for this book is available from the British Library

ISBN: 978-0-12-816398-6

For information on all Elsevier publications

visit our website at https://www.elsevier.com/books-and-journals

Publisher: Susan Dennis

Acquisition Editor: Anneka Hess
Editorial Project Manager: Michael Lutz
Production Project Manager: Vignesh Tamil
Designer: Matthew Limbert
Typeset by VTeX
Contents

PART 1 THEORIES & MODELS

CHAPTER 1 Introduction to two-phase flow . . . . . . . . . . . . . . . . . . . 3
1.1 Flow classifications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
1.1.1 Single-phase flow . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
1.1.2 Gas-particle flow . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
1.1.3 Granular flow . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
1.1.4 Pebble flow . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
1.2 Flow regimes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
1.2.1 Dilute and dense flows . . . . . . . . . . . . . . . . . . . . . . . . 14
1.2.2 Inertial and elastic flows . . . . . . . . . . . . . . . . . . . . . . . 14
1.3 Numerical methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
1.3.1 Methods for single-phase flow . . . . . . . . . . . . . . . . . . . 15
1.3.2 Methods for two-phase flow . . . . . . . . . . . . . . . . . . . . 17
1.4 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20
CHAPTER 2 Discrete particle model . . . . . . . . . . . . . . . . . . . . . . . . . 21
2.1 Spherical particle model . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
2.2 Generalized hard particle model (GHPM) . . . . . . . . . . . . . . . 21
2.2.1 Governing equations . . . . . . . . . . . . . . . . . . . . . . . . . . 21
2.2.2 Consistency with the hard sphere model . . . . . . . . . . . 29
2.2.3 Numerical procedures for implementation . . . . . . . . . . 30
2.2.4 GHPM model validation . . . . . . . . . . . . . . . . . . . . . . . 32
2.2.5 Application in a lifting hopper . . . . . . . . . . . . . . . . . . . 36
2.2.6 Application for the particle-wall collision . . . . . . . . . . 39
2.2.7 Numerical procedure for multiple contacts . . . . . . . . . 44
2.2.8 Simulation test and experimental validation . . . . . . . . . 45
2.2.9 Application in the particle-wall collision . . . . . . . . . . . 50
2.3 SIPHPM model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53
2.3.1 Description of the model basis . . . . . . . . . . . . . . . . . . . 53
2.3.2 Governing equation . . . . . . . . . . . . . . . . . . . . . . . . . . . 58
2.3.3 Equation of motion . . . . . . . . . . . . . . . . . . . . . . . . . . . 58
2.3.4 Numerical procedures and techniques . . . . . . . . . . . . . 59
2.3.5 SIPHPM model validation . . . . . . . . . . . . . . . . . . . . . . 60
2.4 EHPM-DEM model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 64
2.4.1 Extended general hard particle model (EHPM) . . . . . . 64
2.4.2 Extended discrete element method . . . . . . . . . . . . . . . . 67
2.4.3 EHPM-DEM coupling strategy . . . . . . . . . . . . . . . . . . 69
2.4.4 Collision detection strategy . . . . . . . . . . . . . . . . . . . . . 70

v
vi Contents

2.4.5 Governing equations of motion . . . . . . . . . . . . . . . . . . 71

2.4.6 Demonstration and validation of EHPM-DEM . . . . . . . 72
2.4.7 Simulation efficiency . . . . . . . . . . . . . . . . . . . . . . . . . . 76
2.5 Heat transfer extensions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 80
2.5.1 Particle-particle conduction . . . . . . . . . . . . . . . . . . . . . 80
2.6 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 82
CHAPTER 3 Coupled methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 85
3.1 LES-DEM coupled methods . . . . . . . . . . . . . . . . . . . . . . . . . 85
3.1.1 Governing equations . . . . . . . . . . . . . . . . . . . . . . . . . . 85
3.1.2 Discrete element method (DEM) . . . . . . . . . . . . . . . . . 86
3.1.3 Approach A: conventional method . . . . . . . . . . . . . . . . 87
3.1.4 Approach B: smoothed void fraction method . . . . . . . . 88
3.2 DNS-DEM coupled methods . . . . . . . . . . . . . . . . . . . . . . . . . 92
3.2.1 Immersed boundary method . . . . . . . . . . . . . . . . . . . . 92
3.2.2 Point-force method . . . . . . . . . . . . . . . . . . . . . . . . . . . 96
3.3 LBM-DEM coupled methods . . . . . . . . . . . . . . . . . . . . . . . . . 102
3.3.1 Recovery of governing equation . . . . . . . . . . . . . . . . . 102
3.3.2 Coupled with heat transfer . . . . . . . . . . . . . . . . . . . . . . 111
3.3.3 Multiple schemes LBM-IBM-DEM . . . . . . . . . . . . . . . 114
3.4 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 118

PART 2 APPLICATIONS
CHAPTER 4 Application in gas-particle flows . . . . . . . . . . . . . . . . . 123
4.1 Homogeneous turbulence . . . . . . . . . . . . . . . . . . . . . . . . . . . . 123
4.1.1 Governing equations . . . . . . . . . . . . . . . . . . . . . . . . . . 123
4.1.2 Collision rates and statistics . . . . . . . . . . . . . . . . . . . . 127
4.2 Planar jets . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 134
4.2.1 2D case with the heat transfer . . . . . . . . . . . . . . . . . . . 134
4.2.2 3D case with the two-way coupling . . . . . . . . . . . . . . . 144
4.3 Swirling jets . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 148
4.3.1 Vortex breakdown . . . . . . . . . . . . . . . . . . . . . . . . . . . . 148
4.3.2 Coherent oscillation . . . . . . . . . . . . . . . . . . . . . . . . . . 151
4.3.3 Particle-vortex interaction . . . . . . . . . . . . . . . . . . . . . . 158
4.3.4 Four-way coupling . . . . . . . . . . . . . . . . . . . . . . . . . . . 168
4.4 Bubbling fluidized bed . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 173
4.4.1 3D bubbling fluidized testing bed . . . . . . . . . . . . . . . . 173
4.4.2 Pulsed fluidization . . . . . . . . . . . . . . . . . . . . . . . . . . . . 177
4.5 Spouted bed . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 186
4.5.1 CFD-DEM vs. SVFM-based fine LES-DEM . . . . . . . . 190
4.5.2 Particle phase behavior . . . . . . . . . . . . . . . . . . . . . . . . 192
4.5.3 Gas phase behavior . . . . . . . . . . . . . . . . . . . . . . . . . . . 195
4.5.4 Additional remark . . . . . . . . . . . . . . . . . . . . . . . . . . . . 203
4.6 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 204
 Contents vii

CHAPTER 5 Application in granular flows . . . . . . . . . . . . . . . . . . . . 207

5.1 Some functions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 207
5.1.1 Evaluation functions of mixing degree . . . . . . . . . . . . . 207
5.1.2 Evaluation functions of heat transfer degree . . . . . . . . . 214
5.2 Circular drum mixers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 217
5.2.1 Flow pattern and mixing evolution . . . . . . . . . . . . . . . . 217
5.2.2 Dimension analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . 218
5.2.3 Information entropy analysis . . . . . . . . . . . . . . . . . . . . 221
5.2.4 Heat conduction features . . . . . . . . . . . . . . . . . . . . . . . 225
5.3 Wavy drum mixers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 243
5.3.1 Wavy wall configuration . . . . . . . . . . . . . . . . . . . . . . . 243
5.3.2 Analysis and prediction . . . . . . . . . . . . . . . . . . . . . . . . 245
5.3.3 Effects of phase velocity, wave number and amplitude . 247
5.3.4 Driven force analysis . . . . . . . . . . . . . . . . . . . . . . . . . . 256
5.3.5 Heat conduction features . . . . . . . . . . . . . . . . . . . . . . . 261
5.4 Mixing of nonspherical particles . . . . . . . . . . . . . . . . . . . . . . 269
5.4.1 Polygonal particle mixing in 2D drum . . . . . . . . . . . . . 270
5.4.2 Cubic particle mixing in one layer . . . . . . . . . . . . . . . . 287
5.4.3 Tetrahedral particle mixing in one layer tumbler . . . . . 298
5.4.4 Cubic particle mixing in 3D cylinder . . . . . . . . . . . . . . 308
5.5 Hopper discharge . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 320
5.5.1 Geometric features on quasi-static discharge . . . . . . . . 320
5.5.2 Shaken discharge . . . . . . . . . . . . . . . . . . . . . . . . . . . . 329
5.5.3 Discharge of 2D polygonal particles . . . . . . . . . . . . . . 347
5.6 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 357

Bibliography . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 359
Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 373
 CHAPTER

Introduction to two-phase
flow
1
1.1 Flow classifications
In physics and engineering, a flow system is usually characterized by the movements
of the material medium, which can be continuous or discrete, e.g. solids, liquids,
gases, and plasmas. For example, particle flow or granular flow is mainly composed
of solid materials, such as sands, grain, gravels or rocks. A fluid flow usually refers
to the flow motion of liquids, gases, or melted metals, etc.
To characterize the common and essential features of different materials in a flow
system, the concept of ‘phase’ is frequently utilized. A phase is a physically distinc-
tive form of a substance, such as the solid, liquid, and gaseous states of ordinary
matter – also referred to as a macroscopic state [1]. Therefore, a flow composed of
one state of matter is called a single-phase flow, and a flow system with two or more
states of matter is called a two-phase or multiphase flow. It is possible for a system
of a single material to be considered two-phase or multiphase flow, if at least two
distinct dynamical properties of the same material are displayed. For example, a flow
consisting of particles of a single material may be regarded as a two-phase system
if there are fine light particles and heavy inertia particles as two components with
clearly different dynamical behavior. In conclusion, a phase is a continuous or dis-
cretized system characterized by the same physical properties, mechanical states, and
similar dynamical behaviors.

1.1.1 Single-phase flow

In this section, the challenges in modeling of single-phase flows are discussed. One of
the well-known fundamental challenges, which have puzzled the scientists and engi-
neers over hundreds of years, is turbulence. Turbulence is frequently described as ‘ed-
dies’ of varying scales defined in the Kolmogorov’s theory in 1941. Typical features
include irregularity, diffusivity, rotationality and dissipation. Conventional investi-
gation of turbulence was mainly done in the Eulerian framework [2], where the fluid
motion is depicted by the nonlinear three dimensional Navier–Stokes equation. It can-
not be solved analytically, and current computational algorithms are not supported by
computational technology to solve problems of practical scale and conditions, in par-
ticular at high Reynolds numbers. Recently, Liu et al. [3] proposed to reconsider the
definition of vortex and vorticity to characterize the wall turbulence. They have tried
to clarify the onset mechanism and structure of the turbulence and stated that the vor-
Gas-Particle and Granular Flow Systems. https://doi.org/10.1016/B978-0-12-816398-6.00009-2
Copyright © 2020 Elsevier Inc. All rights reserved.
3
4 CHAPTER 1 Introduction to two-phase flow

ticity overtakes deformation in vortex [4–6]. This implies that vorticity and vortex are
completely different concepts. In general, direction of the vorticity is different from
that of the vortex in three-dimensional vortical flows. The vorticity vector (∇ × V )
can be decomposed into a pure-rotational component (R, the vortex vector named
‘Rortex’ or ‘Liutex’) and a nonrotational asymmetrical-deformation component (S).
They declared that, rather than vortex dynamics, R-dynamics (Rortex dynamics or
Liutex dynamics) should be used to investigate the mechanisms of turbulence. Based
on the new vortex identification method, they revisited the Kolmogorov hypothesis,
and declared that turbulent flow has a unique and deterministic solution, which is
not governed by vortex breakdown, but vortex build-up. This work provides a unique
insight on the mechanism of turbulence. In addition, researchers have also tried to
re-examine turbulence under the Lagrangian framework and great efforts were made
to understand the kinematic properties of a fluid particle, such as acceleration and
velocity [7–13].
Single-phase turbulence has wide applications in life. The turbulent jet is one of
the typical examples of turbulent flows with various applications in industry (e.g., free
jet, impinging jet, cross jet, slot jet and jet array) [14]. The vortex structure, mixing
layers and heat transfer are complex yet very important intrinsic characteristics in jet
flows. For example, in the nonisothermal jets, the temperature can be approximately
viewed as scalars driven by turbulence with self-diffusive characteristics. Turbulence
is intrinsically chaotic or fractal. The fractal measure arises from the fact that the
underlying physics of scalar turbulence seems to be closely related to the self-similar
multiplicative fragmentation processes. The mixing and heat transfer interfaces are
always fractal with a noninteger Hausdorff dimension, which implies the stratified
self-similar structure. Fractal dimension could be very important to understand the
turbulence. For industrial applications, the heat transfer interface of a large fractal di-
mension may be beneficial as it can enhance heat transfer efficiency. It is clearly seen
that the effect of fluid properties and characteristics of boundary conditions could
be very important to enhance mixing and heat transfer, as they may affect the inner
interfaces of heat transfer.
A jet flow imparted with a rotational motion about its axis is called a swirling
flow (Fig. 1.1). Swirling flows are frequently encountered in nature, such as torna-
does, typhoons, and ocean currents, which have been studied extensively over the
past decades. The vortex breakdown, related to the coherent structure of the swirling
flows, is considered the most important phenomenon, to which a lot of research has
been devoted [15–22]. The most remarkable feature of the vortex breakdown is the
abrupt change of the vortex structure with pronounced retardation in the axial flow.
Correspondingly, this results in a divergence of the stream surface along the axis [23].
Thus, vortex breakdown can be depicted by three main characteristics: formation of
an internal stagnation point along the axis, reversing flows in a limited region, and
the sudden expansion of vortex core [24]. The vortex breakdown is considered a ma-
jor coherent structure in swirling jets, as the well-organized fluid motion is at scales
comparable to the flow. However in general, there is no accepted definition for the
coherent structure. One attempt is given by Robinson [25] saying that the coherent
 1.1 Flow classifications 5

FIGURE 1.1
The single-phase turbulence in the Eulerian frame: swirling flow.

structure is a region over which at least one fundamental flow variable (velocity com-
ponent, density, etc.) exhibits significant correlation with itself or another one, in time
and space significantly larger than the smallest scales of the flow in the region. Be-
sides swirling flow, the vortex breakdown is also very important in aerodynamics and
combustion systems. For example, in aeronautics, the vortex breakdown can cause
unsteadiness and poor control of the aircraft [26].
In addition, it is necessary to mention that turbulence is not just random as it ap-
pears. Instead, it has intrinsic coherent structures, which occur in a wide variety of
turbulent flows. It is hoped that the work on coherent structures could lead to effi-
cient methods for prediction of overall features of turbulent flows [27]. The coherent
structure in swirling flows has been extensively studied [28], in areas of spatial struc-
ture [29], swirling combustion [30] and instability [31]. In particular, for power plant
using swirling combustors, the vortex breakdown is extensively studied to stabilize
the flame, as the air and fuel drastically mixes in the breakdown region in a result
of the there-circulation toward the stagnation point along the jet axis. With regard to
the swirl instability, for example, Shtork et al. [28] proposed a distinct modulation of
the combustion process by using hydrodynamic instability. Unsteadiness of the flow
is caused by the processing of the left-handed spiral (helical) vortical structure. Tem-
poral frequency of the helical mode can be modulated by superposition of its pure
rotation and axial translation, weighted with coefficients, which are the azimuthal
and axial wavenumbers, respectively.
6 CHAPTER 1 Introduction to two-phase flow

FIGURE 1.2
The gas-particle flow depicted by Eulerian–Lagrangian approach.

1.1.2 Gas-particle flow

In gas-particle flows, the particle is a small object characterized by finite size, vol-
ume and mass (Fig. 1.2). Material of the particle is not limited to solid, for example,
a liquid droplet is also viewed as a particle. Usually size of the particle has a finite
range that excludes the subatomic scaled particles, such as nucleons and electrons.
Microscopic particles of atoms and molecules are also excluded. Gas-particle flows
concern most of particles at macroscopic scale, such as powders and granular ma-
terials. Therefore, classic mechanics (Newton’s laws) can be used to describe the
motion of a discrete particle. The gas-solid flow is one type of two-phase flow which
frequently appears in natural phenomena and industrial processes. For example, the
formation of sand and dust storms, rainfall and avalanche in nature, the process of
boiling, flotation, fermentation, coating, spray drying, and particle transport and mix-
ing in pneumatic conveying and powder mixing devices.
In gas-particle flows, the fluid-to-particle, particle-to-particle, and particle-to-wall
interactions are the three critical fundamental aspects need to be carefully considered.
The forces and torques imposed by fluid-to-particle, and the feedback forces from
particle-to-fluid (including the interfacial distributions) are the key focuses of fluid-
particle interactions. Accordingly, numerical investigations are classified based on
the type of forces considered. If only the fluid-to-particle forces are considered, it is
called the one-way coupling approach. When both the fluid-to-particle and particle-
to-fluid interactions are considered, it is called the two-way coupling approach. In
addition, the particle-to-particle interaction concerns the collision (contact), frictional
forces and torques experienced between particles, where the effects of influencing
 1.1 Flow classifications 7

factors such as particle size, shape, density, roughness, etc. are considered. When
the particle-to-particle, fluid-to-particle and particle-to-fluid interactions are consid-
ered, the method is called four-way coupling. Such an example is the gas-solid flow
of sufficiently high particle concentrations (e.g., the dense flow region in fluidized
beds), where particle behaviors are not only influenced by the aerodynamic trans-
port of turbulence, but also significantly affected by the inter-particle collisions [32].
In addition, the particle-wall interaction should, at least, include the driving forces
from the walls (either stationary or moving), and the effects of wall geometry and
roughness should also be considered.
In two-way coupling of gas-particle flow, the research has been mainly focused
on two aspects: (1) the modification of turbulence [33–35]; and (2) the characteristics
of particle behavior (e.g., particle distribution, dispersion, clustering and coagulation)
[36–38]. The main characteristics of particle-laden flow are governed by groups of
key parameters, and some of them are elaborated here:
• Size (including particle diameter and the characteristic sizes of the turbulence):
∞
• Integral length scale: le = u1u  0 u u (r)dr, where r is the distance be-
tween two measurements, and u is the fluctuation velocity in the same di-
rection. It should be noted that the largest scale always correlates to the most
energetic eddies which may obtain energy from the main flow. Therefore, the
integral length scale le , defined in terms of the normalized two-point flow ve-
locity correlation, is an indication of the largest scales of eddies and velocity
fluctuation, which implies a low frequency in the energy spectrum.
• Taylor microscale λ, corresponding to the inertial subrange, is the intermediate
scale between the largest and the smallest eddies. Taylor microscale eddy does
not dissipate but transport the energy from the largest to the smallest eddies.
• Kolmogorov length scale η: it is the smallest scale in turbulence spectrum
which corresponds to the range in the viscous sublayer, where energy input
from the inertial subrange is balanced with viscous dissipation. These small
scale eddies, usually of high frequency, are always considered homogeneous
and isotropic.
• Mean free path of particle lp : it is the mean path length of the free motion of
particles. lp may start from a collision event and end at the next collision event.
According to the kinetic theory, lp may be affected by the local concentration
and velocity of particles, as well as the particle sizes.
• Time:
• Particle response time τp : driven by the fluid-to-particle force, a particle may
respond to fluid flow in short or long time duration depending on its physical
properties and the interaction of the particle to local turbulence. Usually, the
velocity difference between the fluid flow and particle motion decays expo-
nentially, e.g. in the form of Ce−τ/τp where τp is the characteristic time scale
of the dynamic response of particle in the fluid, or the relaxation time. It is seen
that the relaxation time τp , representing the time lapse for the difference of the
local velocities between fluid and particle phases, decreases to e−1 (herein e is
the Napierian base) of the initial value.
8 CHAPTER 1 Introduction to two-phase flow

• Mean characteristic collision time τc : in a stationary or quasi-stationary gas-

particle flow, τc is the mean characteristic time interval of the free motion for
a particle between two consecutive collisions. It is the mean time interval for a
particle to move freely starting from prior collision with a partner to the next
collision with a different partner.
• The characteristic time scale of fluid turbulence τf : turbulence integral time
scale can be defined
 ∞ in a similar manner as that of the integral length scale
le by τf = u1u  0 u u (t)dt. This corresponds to the characteristic time of
large eddies. In most engineering applications, the characteristic velocity is al-
ways determined by the practical inflow boundary conditions, therefore, τf can
also be derived from the ratio of the characteristic length le and characteristic
velocity.
• Velocity: including the relative velocity vr between particles, and the relative
velocity between particles and the fluid (v p − uf ). Frequently, a derived char-
acteristic velocity, such as vc = le /τf or vc = Re · ν/dp is included in the analysis
depending on the particular applications.
• Density: including the mass or volumetric loading ṁp and V̇p , the particle density
ρp , the fluids density ρg , and the bulk density ρpB , which is defined as the ratio of
the mass of particles to the volume of the gas-particle mixture.
• Temperature Tc : initial temperature difference is usually used as the characteristic
temperature in heat transfer applications.
As the typical applications of gas-particle flows seldom involve electromagnetic
fields, or are concerned with the molecular scale variables, the aforementioned char-
acteristic variables are sufficient to describe the most common cases.
As a convention, dimensionless forms of above-mentioned parameters are com-
monly used to characterize the gas-solid flow, e.g., dp / le , Reynolds number Reλ =
uλ/ν, particle Reynolds number Rep = |v p − uf |dp /ν, mass loading ml = ṁp /ṁg ,
and Stokes number St = τp /τf . Among such examples, Re is a key parameter for
characterizing the fundamental features of the turbulence, and consequentially it can
be used to estimate the needed computer capacity for capturing all scales of the ed-
dies in direct numerical simulation (∼ Re9/4 for spatial resolution and ∼ Re3/4 for
temporal resolution). For the particle dynamics in turbulence, Stokes number is de-
fined as the ratio of the particle response time τp to the turbulence characteristic time
scale τf . When St  1, particle responds to the local turbulence instantaneously so
that it can be regarded as perfect tracer. On the other hand, when St 1, particle
is hardly driven by the fluid so that its motion is almost independent of that of the
turbulence.
Understanding the modulation of turbulence by particle-vortex interaction is of
both practical and theoretical interest. The mechanism of turbulence modulation is
very complicated as it is related to the various aspects of particles, such as particle
inertia [39,40], finite volume or wake effect [41], mass loading and concentration
[42,43], and inter-particle spacing [44]. The results of turbulence modification may
be different from one case to another; therefore, a large number of two-phase flows
are studied to gain knowledge of turbulence modulation under various conditions
 1.1 Flow classifications 9

(e.g., pipe flow [45], mixing layer [46], rotating channel flow [47], and isotropic
turbulence flow [35,48]). With regard to the dimensionless parameter for evaluating
turbulence modulation, frequently dp / le [33] or Rep [49] is used. For example, Gore
and Crowe [33] suggested to use dp / le , the ratio of particle diameter to the integral
length scale (the most energetic eddy), for the demarcation of the turbulence, where
augmentation (dp / le ≥ 0.1) or attenuation (dp / le < 0.1) of the turbulence flow by
suspended particles can be benchmarked. Using a different approach, Tanaka and
Eaton [50] proposed a novel dimensionless number, i.e., the particle moment number
Pa, to classify the turbulence modulation. The Re-Pa relationship was obtained by
examining a set of 80 experimental measurements, and this relationship was clustered
into three groups representing the attenuation, augmentation and transition regions
with two critical particle momentum numbers.
Besides particle properties and fluid turbulence, inter-particle collision is also of
great importance and needs to be carefully treated. Many of the early works [51,
36,34,35] assumed that the particle diameter is comparable to or less than the Kol-
mogorov length scale η. In this context and assuming dilute condition, inter-particle
collision and the particle size effects are usually neglected. Thus majority of the
early researches neglected the collision dynamics. To extend the regime of dilute
gas-particle flow to intermediate dense flow, particle-particle collision should be con-
sidered, and appropriate collision model is needed. Tsuji [52] proposed a model to
describe the collision of rigid granular particles based on the principle of conserva-
tion of the linear and angular momenta. The energy loss in collision was incorporated
through a restitution coefficient e. This model is called the hard sphere model (HSM),
which can be used to simulate the particle motion and collision deterministically. An-
other approach, the discrete element method (DEM) – a soft sphere model (SSM)
[53] – treats the particle-to-particle collision based upon particle kinetics. The soft
sphere model includes three fundamental inter-particle collision mechanisms: elastic
collision with viscous damping, inter-particle friction, and sliding trends. These mod-
els, together with suitable modifications and extensions, will be discussed extensively
in this book.
When the separation of particle centers is small compared to the smallest eddies in
the fluid, and at the same time, particles follow the fluid motion completely, then par-
ticle collision rates (between small particles in turbulent fluids) approximately depend
on the dimension of particles, rate of energy dissipation and the kinematic viscosity
of the fluid [54]. On the other hand, if particle-turbulence velocities are independent,
in analogy to the kinetic theory, particle collision rate in absence of the external force
should follow the suggestions of [55]. The dependence of inter-particle collision on
turbulence and particle properties was studied through direct numerical simulation
of heavy particle suspension in turbulent flow field [37]. By correlating the collision
frequency over several eddy turnover times, two complicated yet well-known effects
of intermediate Stokes number on particles were identified, i.e. (1) particles tend to
collect in regions of low vorticity (high strain) due to the centrifugal effect; (2) parti-
cle pairs are not as strongly correlated with each other, which leads to an increase of
relative velocity. These two effects cause the collision rates to increase locally.
10 CHAPTER 1 Introduction to two-phase flow

FIGURE 1.3
The granular bed.

1.1.3 Granular flow

A granular system is composed of a large number of granules (e.g. the packed gran-
ules in Fig. 1.3), which are abundant in dissipative interactions between particles due
to deformation and friction in body contacts. In such systems, the random Brownian
motion of particles is always omitted as well as the flow of interstitial gas or liquid
(except for the forces caused by them, e.g. the liquid-bridge force). With wide engi-
neering applications, many scientists and engineers have been interested in granular
systems, and great progress in fundamentals and the underlying mechanisms have
been achieved.
Although generating, transporting and stocking of granules can be traced back to
the era of ancient human civilizations, the scientific study of granular flow started
from the 18th century when the Coulomb’s friction law was proposed. Up to date, the
mechanism of granular flow is still poorly understood. It remains one of the central
challenges for scientists [56]. Because of the incomplete knowledge, major frame-
work is on developing the constitutive models by using phenomenological parameters
in the stress-strain relationship of the granular materials, which is a key practice in
engineering applications. As a result, many studies have been focused on the details
of the striking mechanics of granules, such as nonlinear behaviors, self-organization,
force chain structure, and arch formation.
In engineering, the silos, hoppers, and drums are the most common devices for
processing of granular material. Such examples include the mixing, segregating,
transporting processes inside a hopper in powder engineering, the crushing, grinding,
and blasting processes in mineral and mining engineering, the feeding and coating
22 CHAPTER 2 Discrete particle model

FIGURE 2.1
Sketch of particle-particle collision.

simplify the collision process: (1) The collision is three-dimensional, taking place
between rigid particles of arbitrary shapes. (2) The momenta, including translational
and angular, are conserved during collision. (3) The Coulomb’s law of friction is fol-
lowed when inter-particle sliding motion on the contact surface takes place. (4) In the
normal direction of the contact surface, a normal coefficient of restitution e is defined
as the rate of magnitudes of the post-collisional relative velocity to the pre-collisional
relative velocity between the pair of colliding points (C  and C  ), i.e.

(vα(j ) (C  , t) − vα(i) (C  , t))nα = e(vβ (C  , t) − vβ (C  , t))nβ

(i,0) (j,0)
(2.1)
(j ) (i)
where vα and vβ are velocities of the pair of colliding particles ‘(j )’ and ‘(i)’,
respectively. Herein the Einstein summation convention is used for the vector and
tensor representation, i.e. v = vα eα = vβ eβ . The unit orthonormal coordinates eα are
always omitted for simple representation, and v = vα = vβ . ‘C  ’ and ‘C  ’ denote the
contact points and t is time. The superscript ‘0’ denotes the pre-collisional variables.
Otherwise, it represents a post-collisional variable. ‘nα ’ is the normal direction of the
contact surface on the contact point C  (Fig. 2.1).
Based on the conservation laws of momenta, the relations between the post- and
pre-collisional translational and angular velocities can be expressed as follows
 t0 +δt
m(i) vα(i) = m(i) vα(i,0) + fα(i) dt = m(i) vα(i,0) + pα(i) (2.2)
t0
 t0 +δt
m(j ) vα(j ) = m(j ) vα(j,0) − fα(i) dt = m(j ) vα(j,0) − pα(i) (2.3)
t0
 t0 +δt
(i) (i,0) (−1,i) (i) (i,0) (−1,i) (i)
ωβ = ω β + Iβα αγ λ rγ(i) fλ dt = ωβ + Iβα αγ λ rγ(i) pλ (2.4)
t0
 t0 +δt
(j ) (j,0) (−1,j ) (i) (j,0) (−1,j ) (i)
ωβ = ωβ − Iβα αγ λ rγ(j ) fλ dt = ωβ − Iβα αγ λ rγ(j ) pλ
t0
(2.5)
where vα , ωα , m, Iαβ are the translational velocity vector, angular velocity vector,
(−1)
mass and the inertia tensor of three-dimensional particles, respectively. Iβα is the
Another random document with
no related content on Scribd:
“Honored by fools!” growled Roland.
“Honored by the wisest men in college! Honored by every one! If you
had seen every person in that great crowd on the campus rise when
his name was spoken by the historian——”
“I did see it, and then I got away.”
“Then you were there? But you were not in your place.”
“If I had been, they would have seen that one college man did not
rise when Merriwell’s name was called.”
“And you would have brought on yourself the scorn of every one.
Can’t you see that by his generosity, his fine character, and
manliness, he has risen far above you?”
“No! I see that he has a trick of fooling everybody but me. He can
make his enemies forget that they were once enemies, but I am not
like the others. I want to tell you something, Oll. You think Merriwell
has triumphed, but you are wrong. I am the one who has triumphed,
though no one save myself knows it. Some time Merriwell shall
know, and then he will realize that one of his enemies was more than
a match for him.”
“What do you mean?” asked Oliver, amazed. “Are you crazy?”
“Never mind what I mean, but I speak the truth. I have triumphed,
and Merriwell is my victim. I’ll talk no more about it, so you may as
well close your face.”
And Oliver could get nothing further from his brother.
 CHAPTER VII.
THE END COMES.

The day of graduation came. The sun shone bright and clear on this
great day in the life of Frank Merriwell, but still that feeling of
sadness was lingering in his heart, for he felt that he was bidding
farewell to his dear home.
Frank had competed to be a Townsend teacher, and he had been
chosen one of the fortunate six who were to speak for the DeForest
gold medal.
Thus it happened that he was given little time for thought and little in
which to see his friends, all of whom were eager to be in his
company.
Had he known that the oilskin envelope in his possession contained
nothing but blank paper it is not probable he could have spoken as
brilliantly as he did.
When the speaking was over it was generally conceded that the
handsome medal must go to Merriwell.
The faculty adjourned to the Treasury building, and there Frank was
awarded the splendid prize. Each member of the faculty shook his
hand in turn and spoke some word of praise to him. They looked on
him lovingly, for they knew that he had done more to raise the
standard of college life than any other student in the country.
Frank was on his way to his room when he almost collided with
Roland Packard.
Packard had been drinking heavily, and he stopped, his lip curling in
a scornful sneer.
“You think you’re it, Merriwell,” he said, in a tone of great contempt;
“but, if you only knew it, you are the biggest fool alive.”
Frank had no desire to exchange words with the fellow.
“You’re drunk, Packard,” he said quietly.
“You’re a liar, Merriwell!” snarled Packard, who seemed not to have a
single remnant of reason left.
Frank was not in the habit of taking the lie from anybody, but now,
seeing Packard’s arm in a sling, he did not heed the fellow’s insult.
“Your friends think you’re a great gun,” Roland went on; “but you
really are mighty small potatoes. Won the DeForest prize, did you?
Well, you may have to pawn it soon to get bread to keep you from
starving!”
This did not have the effect Roland had fancied it might, which
angered him to a still further expression of rage.
“Oh, you’re mighty cool; but you won’t be so cool when you find
you’re a beggar! And you are! I know what I’m talking about. You will
find it out in time, and I want to tell you now that it is I—I, Roland
Packard, whom you despise, who has made you a beggar! Don’t
forget it!”
He wheeled and walked swiftly away.
Frank stood still and looked after the fellow.
“I wonder what he meant,” Merry muttered, a feeling of uneasiness in
his breast. “Is he plumb daffy? I know he’s pretty drunk, but still it
seems that he must have some reason left.”
Frank was troubled despite himself, and he hurried to his room,
where he made sure the oilskin envelope was still safe in his
possession.
Packard had hurried away to drink still more. Already he was half-
crazed by liquor, but he felt consumed by a burning fire that called
for more, more, more.
The afternoon of graduation-day came and saw all graduating
students in caps and gowns, headed by the faculty, likewise garbed,
march to the music of a band out of the campus and down Elm
Street to the green, which they crossed, turning up Chapel Street to
Vanderbilt. The gates of Vanderbilt are opened but once a year,
always on this occasion, and through the gates they marched, under
the arch and across the campus. The chapel was entered, and then
came the last solemn ceremony of conferring the degrees.
Frank thrilled when he stood up to receive his sheepskin. There was
a choking in his throat, his sensation was a mingled feeling of joy
and sorrow that was like exquisite pain. His face was pale as marble.
When the certificate was placed in his hand he felt that it was the
document that divorced him from dear old Yale, and he sat down
with his teeth clenched to hold back the moan that sought vent.
It was over!
That afternoon a man was seen reeling over the Barnesville bridge.
He was intoxicated, and he seemed to fancy he was pursued by an
enemy or enemies who sought his life. Filled with mad terror, he
climbed upon the railing not far from the eastern end of the bridge
and flung himself headlong into the river.
Several persons had seen this crazy act, and they rushed to rescue
him, if possible. Two men pulled out in a boat toward the spot where
he had last been seen. As they pulled he rose to the surface, made a
few feeble splashes, and sank.
One of the men stripped off his coat and plunged in. He brought the
drowning fellow up, helped the other man get him into the boat,
crawled in himself, and they pulled ashore.
On the shore men worked nearly an hour over the poor wretch, but
all their efforts were unavailing. He was dead. In his pocket they
found some letters, which told them he was a student and that his
name was Roland Packard.
And thus it came about that in the pocket of his dead brother Oliver
Packard found another envelope that looked exactly like the one
Roland had snatched from Merriwell. He was astonished and
puzzled, but he took it to Merriwell.
“One of them must contain the message, Merriwell,” said Oliver,
whose face was marked with deep sorrow.
“To-morrow will tell,” said Frank, “for then I will open them both.” He
took Oliver’s hand. “I am very sorry, Packard,” he said.
“It is for the best,” declared Oliver; but his chin quivered as he turned
away.
 CHAPTER VIII.
THE MESSAGE STOLEN AGAIN.

“The time has come!”

The words came from the lips of Frank Merriwell, who was standing
beside a small table in a room of one of New York’s big hotels. In his
hand he held the two oilskin envelopes. Across each envelope had
been written:
“To Frank Merriwell; to be opened the day after he graduates from
Yale.”
Frank had studied the writing on those envelopes, and he was
convinced that the words on one had been imitated and copied from
the other.
Bart Hodge was Merry’s companion, sitting near and showing no
small amount of interest in the singular envelopes.
“Which contains the message?” was the question that came from
Bart’s lips.
“That is a conundrum,” admitted Frank, as he gazed from one to the
other.
“This is the one Oliver Packard returned that night the old grads
were celebrating on Osborne corner.”
“Which one is that, the original or the fake?”
“The original.”
“Then what do you make of it?”
“I believe it does not contain the message. I believe the original
envelope was opened by Roland Packard.”
“Why did he do that?”
“I don’t know, unless he expected he would have to give up
something and was determined to hang on to the real message. I am
convinced that there was somebody behind Roland Packard. He was
not working on his own hook. The messenger was pursued all the
way from Colorado to New Haven by a man who seemed
determined to do him injury. That man failed, but is it not possible he
instigated the action of Roland Packard?”
“And you think the stranger employed him to get hold of the
message?”
“I have arrived at that belief.”
“Still, that does not explain the fake envelope.”
“It seems to me that Roland Packard’s curiosity was aroused and he
determined to find out what the original envelope contained. He
opened it. In fact, having studied and examined this envelope
closely, I think I can detect indications that it has been broken open.”
“Then it is likely that Oliver Packard did not restore to you the
message, after all.”
“Not in this first envelope, but you know he brought me this other,
which was taken from the body of his dead brother.”
“Then it is possible that the second envelope is the one that contains
the message.”
“Yes,” nodded Frank. “I almost dread to open it, although the time to
do so has come. Something seems to whisper that it contains a
great surprise for me.”
Frank sat down beside the table, and, with a firm hand, tore open the
envelope he regarded as the original. An exclamation escaped his
lips as he drew forth the contents.
“Look, Bart!” he cried. “I was right! Nothing but blank paper!”
He held the unsoiled sheets up before the eyes of his almost
breathless companion.
“By Jove! you were right!” said Hodge. “You have a way of figuring
things out correctly, Merriwell. The other envelope must contain the
message.”
But, strange to say, Merriwell seemed to hesitate again.
“What if it should not!” he muttered. “What if that also contains
nothing but blank paper!”
“But it must contain the message!” exclaimed Bart.
“Why?”
“Because—because the message was not in this one.”
“A poor reason, Bart. It’s likely this envelope was fixed to deceive the
man who employed Roland Packard to secure the message. I
presume that man offered Packard money to get the message and
turn it over to him. Packard’s curiosity was aroused, and he decided
to find out what the message contained, which led him to remove it
from the envelope. Then he fixed up the original envelope to deceive
the man who had paid him to do the crooked work, but his brother
took it from him in the fight. Following that it is likely that he fixed up
this other envelope for the purpose of fooling his rascally employer.
In such a case, it is almost certain that envelope No. 2 contains
blank paper, the same as the first.”
“Open it!” panted Hodge.
“That will settle it,” said Frank, as he did so.
Bart was rigid as a marble image as Merry drew the contents of the
envelope forth.
From Frank’s lips came a sigh of satisfaction.
“It is the message!” he said.
Had he not been so preoccupied, so absorbed, Frank Merriwell
would have heard the slight rustling sound in the alcove bedroom
behind him. In times of expected danger his alertness was
something remarkable, but just now his mind was concentrated on
the mysterious message which he had taken from the envelope.
Nor did Bart hear anything to arouse his suspicions.
A slight breeze came through the open bedroom window, and gently
stirred the portières behind Frank’s back.
Merriwell’s face grew very pale as he read the opening words of the
message, and his watching companion knew something had
produced a profound effect on him.
“What is it?” Bart was compelled to ask.
“It is from my father, as I believed,” said Merriwell, plainly making an
effort to steady his voice. “I have read nothing but the opening
sentence, but this is what it says:
“‘This, my son, is the confession of your father, who, near to the point
of death and beyond all hope of recovery, is lying in the cabin of
Juan Delores, near Urmiston, which is about fifty miles from
Denver.’”
“Great Scott!” exclaimed Hodge. “Your father dying?”
“Dead by this time, it is likely,” came sadly from Frank’s lips. “And I
not near in his last moments!”
The expression of regret and grief on Frank’s face was sincere and
profound.
“Too bad!” muttered Bart. “But he always was such a strange man!”
“Strange, indeed,” nodded Frank. “I knew little of his life after he
went to seek his fortune amid the mines, save that part which is
closely connected with his fight against his great enemy, Santenel.
He told me that portion of it, but concerning the rest he has said little
or nothing.”
“This may throw light upon it. He calls it a confession.”
“And the fact that he has called it that makes me hesitate once more
about reading. But it must be done.”
Again Merry lifted the message to read.
Over his shoulder darted a hand that snatched the message from his
grasp!
At the same moment, uttering a cry of warning, Bart Hodge sprang to
his feet, pointing toward the parted portières behind Merriwell.
Merry shot to his feet like a flash, but he was barely in time to see a
man disappearing between the portières.
A second time had the precious message been snatched from his
fingers.
“Stop him!” shouted Hodge.
Merry was first to leap between the portières, and yet he was barely
in time to see a man disappearing through a window that led out
upon a fire-escape.
A single glimpse of the man’s face Merriwell obtained as he plunged
after him. He saw him entering the open window of an adjoining
room, the fire-escape running from one window to the other.
At a single bound Frank reached the other window and followed the
man into the room. The fugitive was passing out through a door that
led into the hall as Merry jumped in by the window.
Toward that door bounded Merry. It was slammed in his face.
It had a spring-lock, and for a moment it bothered Frank, who was
compelled to pause to open it. By that time Hodge had reached the
window of the room, into which he looked in great surprise, seeing
that Merry was there alone.
“Where is the——” Frank heard no more of Bart’s question, for he
tore open the door and leaped out into the corridor.
The fugitive had disappeared.
Frank went dashing along the passage, looking for the man, but
seeing nothing of him. The fellow had disappeared in a most
remarkable manner after leaving the room.
“Search, Hodge!” called Merry, and Bart joined in the hunt.
But though they searched everywhere, they found nothing of the
man they were after. The hotel was aroused. The clerk in the office
was notified, and he sent the hotel detective to join in the search.
But, after an hour of hunting, the searchers were forced to give up,
as the unknown thief had not been found.
Then Merry went to the office and took a look at the register to find
out who had occupied the room next to his—the one through which
the desperate rascal had made good his retreat from the fire-escape.
The name on the register was “Anton Mescal, Fair Play, Col.”
“Fair Play!” muttered Hodge, who was looking over Frank’s shoulder.
“What does a scoundrel like that know of fair play?”
Frank asked the clerk if he could give a description of Mescal.
“He is slender, looks like a Spaniard, and has a small, pointed, black
mustache,” was the answer. “I do not remember how he was
dressed, so his clothes must have been fairly within the style.”
“That’s the man!” exclaimed Hodge. “I saw his face, and the
description fits.”
Frank nodded.
“I believe Mescal is the man,” he said. “I will give one thousand
dollars for his capture and the restoration to me of the document
which he snatched from my hands.”
The clerk looked at Merry, as if doubting his ability to pay such a
sum; but the young Yale graduate was taking a small roll of bills from
his pocket. From the roll he drew off two five-hundred-dollar bills,
which he handed to the cashier, who stood near the clerk.
“The money is to be paid to the person or persons who capture or
cause to be captured the thief who stole the document from me, in
case it is restored to my hands,” said Merriwell quietly. “You are to
enlist the services of the regular police and do everything in your
power.”
“The police have been called already,” said the clerk. “I telephoned
the nearest station immediately, and two officers appeared very
shortly. They have been guarding the entrances to the hotel, while
the regular house detectives have been searching. I suspected this
Mescal and gave an accurate description of him to the policemen.
They have not stopped him as yet.”
“Only two officers on guard!” exclaimed Frank. “Yet there is a front
and back entrance, and one through your barber’s shop and by the
way of the bar. Mr. Mescal is out of the hotel by this time.”
“We have done everything we could” declared the clerk.
Frank turned away.
“The message is lost, Bart,” he said.
“Lost?” said Bart, astonished that Frank should give up so easily.
“Yes,” Merry nodded, his face wearing a grim expression.
Hodge was trembling with rage at the outcome.
“It’s an infernal shame!” he hissed. “Merriwell, you must——”
Frank’s hand gripped his arm.
“Come!” said Merry’s voice, still calm and restrained.
Together they went to the nearest police-station, where Frank told
his story to the sergeant in charge, repeating his offer for the arrest
of the thief and the restoration of the message. He was told that
everything possible should be done, and with that promise he was
compelled to be satisfied.
Frank scarcely spoke as they returned to the hotel. Bart wiped the
perspiration from his face and said things to himself.
In his room Merry sat quite still for some time, the look on his face
indicating that he was in deep thought.
Bart did not venture to break in upon his meditations. To Hodge this
second loss of the message, at the moment when Merry had begun
to read it, was something to throw him into a perfect tempest of rage;
but Frank had shown that he was master of his temper.
Bart knew Merry was thoughtfully considering the situation and
studying over it in view of the proper course to pursue. After half an
hour he quietly said:
“That is what I’ll do.”
“What is it?” asked Bart, unable to repress his curiosity longer. “What
have you decided to do?”
“I believe there is not one chance in a thousand that the man who
snatched that message will be captured before he can get out of
New York, and this has led me to decide on a course of action. In the
single sentence that I read my father said that he was at the cabin of
Juan Delores, near Urmiston, which is about fifty miles from Denver.
I shall wait here until to-morrow. If the police have not made a
capture by that time, I shall leave New York.”
“Whither bound?”
“For the cabin of Juan Delores, near Urmiston, Colorado. I am going
to find out the truth, if possible. There is a mystery to be solved, and
I mean to solve it. Bart!”
“Frank!”
“Are you with me?”
Merry had risen. Hodge leaped to his feet. Their hands met, as Bart
exclaimed:
“To the end, through thick and thin!”
 CHAPTER IX.
THE OLD INDIAN.

Before them lay the mighty Rockies, rising range on range, till their
glittering, snow-capped summits pressed the sky. Wild and
picturesque and awe-inspiring was the scene. They were in the foot-
hills, and the country was rough and broken.
Frank had drawn rein at the mouth of what seemed to be a small
valley. He was covered with dust, and the hardy mustang he
bestrode showed signs of weariness.
Merriwell was clothed to rough it, having exchanged the garments of
the cities and towns for those more suited to the latter stages of his
search for the cabin of Juan Delores. On his head was a wide-
brimmed felt hat, and he wore a woolen shirt, with a side collar and a
flowing tie, a cartridge-belt about his waist, and leather leggings
covered his trousers nearly to his thighs. There were spurs on the
heels of his boots. His coat he had stripped off, for the day was
warm to an uncomfortable degree.
A Winchester repeating rifle was slung at the pommel of Merry’s
saddle, and a pair of long-barreled revolvers rested in the holsters on
his hips. Taken altogether, he looked like a young man who had
made preparations for almost anything he might encounter.
Bart Hodge, similarly mounted and dressed, had drawn up beside
Frank.
Despite their attire, there was something in the appearance of the
two young men that marked them as belonging to “the tenderfoot
breed.” In other words, the experienced eye would have discovered
at a glance that they were Easterners.
A cool breeze came down the valley, bearing with it a pleasant odor
of wild growing things.
The faces of both lads, lately fresh from college, had been burned
and blistered by the hot suns and searing winds.
“It’s remarkable,” said Frank, “that the people at Urmiston know
Delores, know he lives somewhere in this vicinity, yet not one of
them could give us accurate directions to reach his cabin.”
“Hanged remarkable!” growled Bart. “This is the third day we have
spent in hunting for his old place, and we’ve not even found a clue to
it.”
Merry nodded, frowning beneath the wide brim of his hat.
“We may have passed and repassed it,” he said. “There are plenty of
places where cabins could be hidden in these valleys.”
“That’s right. What are we to do?”
“Keep on hunting.”
“It’s rather tiresome.”
“I shall stick to it till I find the cabin of Delores, if it takes a year!”
exclaimed Frank grimly.
Bart knew he would do exactly as he said.
“Perhaps we may be disappointed when we do find it.”
“At least, I should be able to learn if my father is dead, and where he
is buried.”
“But the message——”
“I have hopes that I may learn the secret of that, also. It may be that
he did not trust it alone to that one document.”
“It’s getting late. What are we to do now? Shall we explore this valley
to-night, or wait till morning?”
Little of the valley could be seen through the narrow pass, and that
little seemed to promise that it led onward far into the hills. After a
moment Frank answered:
“We’ll ride forward and see if we can get a look into it.”
He started onward, and Bart followed, but they had proceeded only a
short distance when they were startled to see, sitting on a boulder at
one side of the pass, a strange figure. At first it was hard to make out
whether it was man or woman, but, as they drew nearer, it
straightened up and revealed, peering from the folds of a dirty red
blanket, the wrinkled and gnarled face of an old Indian. A pair of
beady black eyes were steadily regarding the two young men.
“Watch him, Merry,” cautioned Bart, in a low tone. “These half-
civilized red dogs are treacherous.”
The Indian did not stir as they approached. Beside him, leaning
against the boulder, was a handsome rifle. He did not touch the
weapon.
“Hello, chief,” said Frank, addressing the old man in a manner he
knew was flattering to some redskins, as he drew up.
“How, how,” grunted the old fellow, in answer.
“Are you acquainted in this vicinity?”
“Ak-waint?” said the old man. “No savvy.”
“Are you familiar with the country?”
“Fam-mil? What him?”
“Have you been all round every place here?” asked Merry, with a
sweep of his arm, using the simplest words he could command.
“Heap been all over,” was the assurance.
“Know Juan Delores?”
“Him don’t live round here.”
The answer was prompt enough—a trifle too prompt, Frank fancied.
“Doesn’t?” said Merry. “Where does he live?”
“Heap long way off there,” and the redskin pointed to the north.
“Are you sure?”
“Heap sure.”
“How far? How many miles?”
“Two time ten.”
“Twenty?”
The old fellow grunted an affirmative.
“Do you know the way to his place?”
Another affirmative grunt.
“Can you guide us there?”
“No time.”
“We will pay you well.”
“No time.”
“I will give you fifty dollars to guide us to the cabin of Juan Delores.”
“No time.”
“A hundred dollars.”
“No time.”
“Confound him!” growled Hodge angrily. “Money is no object to him.
It’s likely he doesn’t know the value of money. Now, if you had a
quart of whisky to offer him, Merriwell, you might get him to do the
job.”
“I will give you a new blanket and a rifle,” promised Merry.
“Got blanket an’ rifle,” said the old Indian.
“I will give you a good horse.”
“Got heap good horse.”
“What haven’t you got that you want?”
“No want nothin’.”
“Will you tell us how to get to the cabin of Delores?”
“Go there two time ten mile, find stream, go up him to spring, take
trail from spring; it make you come to where Juan he live.”
Merriwell was not at all satisfied with these directions. There was
something in the manner of the old redskin that seemed to arouse
his suspicions and make him feel that he was being deceived. Of a
sudden Frank asked:
“Who lives in this valley?”
The old man shook his head.
“No know,” he said. “Wolf, bear, mebbe.”
“That’s not what I mean. Is there a white man who lives in this
valley?”
Again a shake of the head.
“Wolf, bear, that all. No; big mount’n-lion—him there. Him kill hunter
—one, two, t’ree, four hunter—what come for him. Him vely bad lion
—heap bad.”
Frank was watching the man closely.
“That’s just what I’m looking for!” he exclaimed, as if delighted. “I
want to shoot a mountain-lion.”
“You no can shoot him. Big hunter try—no do it. Him kill you heap
quick, you go in there.”
“He is trying to frighten us so we’ll not go into the valley,” thought
Frank. Aloud he said:
“That’s all right; I’ll take chances. I reckon the two of us will be too
much for Mr. Lion.”
“White boy much foolish,” declared the old redskin grimly. “Make big
supper for lion. Lion him like white man for supper.”
“And I’ll have the pelt of that lion just as sure as I live,” said Merry, as
if in sudden determination. “Come on, Bart!”
The old Indian rose quickly as they were about to start forward.
“Stop!” he cried. “Ole Joe Crowfoot him tell you truth. If you go in
there you never come back some more. Ole Joe Crowfoot him good
Injun—him like white man heap much. No want to see um hurt. Tell
um to stay back.”
The old savage seemed deeply in earnest now, but that earnestness
was something that added to Frank’s suspicions and made him all
the more determined to go on.
“That’s all right,” said Merry, with a grim smile. “It’s kind of you to
take so much interest in us, but we’re going after your heap bad lion,
and we’ll have his pelt.”
“Night come soon,” said the Indian, with a motion toward the range
on range of mountains rising to the westward. “Then lion him crouch
and spring. Him git you quick.”
“We’ll see. If you wait round here long enough we’ll show you the
pelt of your bad lion when we come back.”
“No come back,” declared Old Joe Crowfoot, solemnly. “No see you
some more. By-by.”
An expression of deep sadness and regret was on his wrinkled old
face as he uttered the words. Merry laughed lightly, and they rode
past him and headed onward into the valley.
“He was very anxious to stop us,” said Hodge.
“That’s right,” nodded Frank. “He was altogether too anxious. As
soon as I tumbled to that I decided to take a look into the valley. Do
you know, we stumbled on the entrance to this valley by accident. I
fancy we might search a week for it, if we were to go away now,
without finding it.”
“I was thinking of that,” said Bart. “It might puzzle us to find it again.
Perhaps that old duffer was counting on that. Those red dogs are
treacherous, and——”
They heard a sharp cry behind them. Whirling in the saddle, Frank
saw the old Indian standing with the butt of his rifle pressed against
his shoulder.
The muzzle on the rifle was turned directly toward Frank, and plainly
the redskin was on the point of pressing the trigger.
Frank knew he was in deadly peril, and he would have attempted to
fling himself from the saddle but for something else he saw.
On a mass of jagged rocks behind the Indian and about twenty feet
above his head had appeared a boy. Not over thirteen years of age
was the lad, whose curly, dark hair fell upon his shoulders. He was
dressed in fanciful garments, like those worn by a young Mexican
lad, and the bright colors of his clothes made him a picturesque
figure.
Plainly it was from his lips that the cry had issued.
In his hand the boy held a stone as large as a man’s fist, and even
as Merry turned he hurled the stone. Straight through the air whizzed
the missile, striking the barrel of the old Indian’s rifle.
Smoke belched from the muzzle of the weapon and the crags flung
back the sound of the report, but the bullet flew wild.
Frank Merriwell’s life had been saved by the stone thrown by the
strange boy.
With an exclamation of rage, Hodge snatched up his rifle and reined
his mount round to take a shot at the redskin, who had wheeled
instantly and was clambering up the rocks toward the boy, as if bent
on murder.
“Soak him, Merry!” panted Bart.
Frank’s first impulse was to shoot, but he quickly saw that he was in
no further danger just then, and he had no desire to shed human
blood unless compelled to do so.
Bart’s rifle rose, but Merry thrust the muzzle aside just as the
weapon spoke, and the bullet flattened on the rocks.
“Why did you do that?” roared Hodge, in amazement and anger.
“Can’t you see! That red devil is going to murder the kid!”
It did seem that the Indian meant the boy harm, and Merry shouted:
“If you put a hand on that boy I’ll bore you!”
At the same time he held his own rifle ready for instant use.